I'm using Agilent 4v44k v4 chips and trying to analyse the data with Robin.

I can load the txt files, view the quality characteristics and create groups and metagroups in the experimental design window.

Everything looks fine as far as the signal and hybridisation go.

However, once I click "next" in the experimenrtal design window, the "running analysis" starts and never finishes.

I've tested the software with some old Affy data and it works fine.

Any ideas?

Hello,

thanks for your interest in using Robin. In order to help you it would be helpful to get as many details as possible;
e.g. What operating system are you running Robin on? Which Java version is
installed? When opening the task manager (by pressing CTRL-ALT-DEL) while
the main analysis is running infinitely, do you still see active R processes producing
processor load? What is the data volume of your experiment? How much RAM memory
does the computer you started the analysis on have? The analysis might take up
to several hours for large datasets, depending on the computer hardware used.

best greetings,
Marc

Hi,

Operating system: Windows XP Pro, SP3.

Java 6 Update 19 (build 1.6.0_19-b04)

Computer: Intel Core 2Duo CPU, E6550 @ 2.33GHz, 1.95 GB RAM

Data volume: Maximum tested - 12 chips, 4 groups. Minimum tested - 6 chips, 2 groups.

Experimental design - Group 1 vs Group 2, Group 3 vs Group 4

Task Manager after clicking "next" at Experimental design stage: Processes - System Idle Processes 99%, taskmgr.exe 1%.

(I also checked the Task Manager while running an affy analysis and it showed Rterm.exe running at 50%)

I did notice the difference between the affy and agilent analyis at this stage - there were no normalisation or p value options in the left hand panel.

I also tested the program on a different PC and had the same issue. Seems like the software is not running after the experimental design phase?

Thanks,
Matt

Hi Matt,

judging by your system's specifications everything should be fine. Memory does not seem
to be the issue since loading the data to run the quality checks works fine (right?). So the
problem must occur during the statistical analysis. The easiest way for me to find out what's
going wrong would be if you sent me a sample of your data (e.g. the minimum tested set that
does show the problem) so that i can reproduce it and analyse the problem. Of course your data
will be treated confidentially and i will only use it to track down the problem and delete it afterwards.
If you want to contact me directly please email to : lohse@mpimp-golm.mpg.de

best greetings,
Marc

Hi,

the behaviour was caused by a bug that occasionally caused the process to hang when
analysing some datasets. New download packages containing a fix for this (and other
issues) will be released latest next week.

Thanks for your patience,
Marc